moonshotai/kimi-k3

Online Chat

月之暗面

Publish time

2026/7/17

Model Series

Kimi

Input type

Output type

Input Price

¥20 / 1M tokens

Output Price

¥100 / 1M tokens

Cache Write Price

¥25 / 1M tokens

Cache Read Price

¥2 / 1M tokens

Context Window

1,000,000

Max Output Length

1,000,000

Kimi K3 是 Kimi 迄今能力最强的旗舰模型，拥有 2.8 万亿参数，基于 KDA 混合线性注意力机制（Kimi Delta Attention）和注意力残差（Attention Residuals）技术构建，原生支持视觉理解，并拥有 100 万 token 上下文窗口。它是全球首个开源的 3 万亿级别模型，面向长程编程、知识工作和推理等前沿智能场景而设计。

Providers for moonshotai/kimi-k3

Zhinao API routes requests to the best-fit provider and automatically fails over to the one with highest availability.

Sort by

paratera并行科技

国内

TTFT

20.00s

Throughput

No data

Uptime

50.00%

Provider Model

paratera/moonshotai/kimi-k3

Supported Parameters

temperaturetop_ptop_k

Recent Uptime

10月11日 11 PM50.00%

Reasoning

Toggleable

Supported Response Formats

OpenAI Chat Completions

Total Context

1,000,000

Max Output

1,000,000

Input Price

¥20 / 1M tokens

Output Price

¥100 / 1M tokens

Cache Write

¥25 / 1M tokens

Cache Read

¥2 / 1M tokens

tokenmall-cdtx

国内

TTFT

67.66s

Throughput

43.68tps

Uptime

100.00%

Provider Model

tokenmall-cdtx/moonshotai/kimi-k3

Supported Parameters

temperaturetop_ptop_k

Recent Uptime

10月11日 11 PM100.00%

Reasoning

Toggleable

Supported Response Formats

OpenAI Chat Completions

Total Context

1,000,000

Max Output

1,000,000

Input Price-40%

¥12 / 1M tokens

Output Price-40%

¥60 / 1M tokens

Cache Write-40%

¥15 / 1M tokens

Cache Read-40%

¥1.2 / 1M tokens

tokenmall-jdy

国内

TTFT

6.78s

Throughput

26.30tps

Uptime

100.00%

Provider Model

jdcloud/moonshotai/kimi-k3

Supported Parameters

temperaturetop_ptop_k

Recent Uptime

10月11日 11 PM100.00%

Reasoning

Toggleable

Supported Response Formats

OpenAI Chat Completions

Total Context

1,000,000

Max Output

1,000,000

Input Price

¥20 / 1M tokens

Output Price

¥100 / 1M tokens

Cache Write

¥25 / 1M tokens

Cache Read

¥2 / 1M tokens

tokenmall-zyld

国内

TTFT

20.13s

Throughput

16.58tps

Uptime

100.00%

Provider Model

tokenmall-zyld/moonshotai/kimi-k3

Supported Parameters

temperaturetop_ptop_k

Recent Uptime

10月11日 11 PM100.00%

Reasoning

Toggleable

Supported Response Formats

OpenAI Chat CompletionsOpenAI ResponsesAnthropic Messages

Total Context

1,000,000

Max Output

1,000,000

Input Price-43%

¥11.4 / 1M tokens

Output Price-43%

¥57 / 1M tokens

Cache Write-43%

¥14.25 / 1M tokens

Cache Read-43%

¥1.14 / 1M tokens

月

月之暗面

国内

TTFT

12.66s

Throughput

23.93tps

Uptime

100.00%

Provider Model

moonshot/moonshotai/kimi-k3

Supported Parameters

temperaturetop_ptop_k

Recent Uptime

10月11日 11 PM100.00%

Reasoning

Toggleable

Supported Response Formats

OpenAI Chat CompletionsAnthropic Messages

Total Context

1,000,000

Max Output

1,000,000

Input Price

¥20 / 1M tokens

Output Price

¥100 / 1M tokens

Cache Write

¥25 / 1M tokens

Cache Read

¥2 / 1M tokens

火

火山引擎

国内

TTFT

10.58s

Throughput

18.59tps

Uptime

100.00%

Provider Model

volcengine/moonshotai/kimi-k3

Supported Parameters

temperaturetop_ptop_k

Recent Uptime

10月11日 11 PM100.00%

Reasoning

Toggleable

Supported Response Formats

OpenAI Chat CompletionsOpenAI Responses

Total Context

1,000,000

Max Output

1,000,000

Input Price

¥20 / 1M tokens

Output Price

¥100 / 1M tokens

Cache Write

¥25 / 1M tokens

Cache Read

¥2 / 1M tokens

百

百度文心

国内

TTFT

26.25s

Throughput

14.44tps

Uptime

100.00%

Provider Model

baidu/moonshotai/kimi-k3

Supported Parameters

temperaturetop_ptop_k

Recent Uptime

10月11日 11 PM100.00%

Reasoning

Toggleable

Supported Response Formats

OpenAI Chat Completions

Total Context

1,000,000

Max Output

1,000,000

Input Price-10%

¥18 / 1M tokens

Output Price-10%

¥90 / 1M tokens

Cache Write-10%

¥22.5 / 1M tokens

Cache Read-10%

¥1.8 / 1M tokens

通

通义千问

国内

TTFT

7.11s

Throughput

31.90tps

Uptime

100.00%

Provider Model

alibaba/moonshotai/kimi-k3-2

Supported Parameters

temperaturetop_ptop_k

Recent Uptime

10月11日 10 PM100.00%

Reasoning

Toggleable

Supported Response Formats

OpenAI Chat CompletionsAnthropic Messages

Total Context

1,000,000

Max Output

1,000,000

Input Price

¥20 / 1M tokens

Output Price

¥100 / 1M tokens

Cache Write

¥25 / 1M tokens

Cache Read

¥2 / 1M tokens

通

通义千问

国内

TTFT

9.34s

Throughput

20.90tps

Uptime

100.00%

Provider Model

alibaba/moonshotai/kimi-k3

Supported Parameters

temperaturetop_ptop_k

Recent Uptime

10月11日 11 PM100.00%

Reasoning

Toggleable

Supported Response Formats

OpenAI Chat Completions

Total Context

1,000,000

Max Output

1,000,000

Input Price

¥20 / 1M tokens

Output Price

¥100 / 1M tokens

Cache Write

¥25 / 1M tokens

Cache Read

¥2 / 1M tokens

Performance for moonshotai/kimi-k3

Compare different providers across Zhinao API

Throughput

21.77 tok/s

TTFT

19.98 s

Uptime for moonshotai/kimi-k3

Uptime for moonshotai/kimi-k3 across all providers

Sample code and API for moonshotai/kimi-k3

Get API Key

Zhinao API normalizes requests and responses across providers for you

View full docs

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.360.cn/v1",
  apiKey: process.env.ZHINAO_API_KEY,
});

const response = await client.chat.completions.create({
  model: "moonshotai/kimi-k3",
  messages: [
    { role: "user", content: "Hello, how are you?" }
  ],
  temperature: 0.7,
  max_tokens: 1000,
});

console.log(response.choices[0].message.content);

moonshotai/kimi-k3

Online Chat

月之暗面

Publish time

2026/7/17

Model Series

Kimi

Input type

Output type

Input Price

¥20 / 1M tokens

Output Price

¥100 / 1M tokens

Cache Write Price

¥25 / 1M tokens

Cache Read Price

¥2 / 1M tokens

Context Window

1,000,000

Max Output Length

1,000,000

Providers for moonshotai/kimi-k3

Zhinao API routes requests to the best-fit provider and automatically fails over to the one with highest availability.

Sort by

paratera并行科技

国内

TTFT

20.00s

Throughput

No data

Uptime

50.00%

Provider Model

paratera/moonshotai/kimi-k3

Supported Parameters

temperaturetop_ptop_k

Recent Uptime

10月11日 11 PM50.00%

Reasoning

Toggleable

Supported Response Formats

OpenAI Chat Completions

Total Context

1,000,000

Max Output

1,000,000

Input Price

¥20 / 1M tokens

Output Price

¥100 / 1M tokens

Cache Write

¥25 / 1M tokens

Cache Read

¥2 / 1M tokens

tokenmall-cdtx

国内

TTFT

67.66s

Throughput

43.68tps

Uptime

100.00%

Provider Model

tokenmall-cdtx/moonshotai/kimi-k3

Supported Parameters

temperaturetop_ptop_k

Recent Uptime

10月11日 11 PM100.00%

Reasoning

Toggleable

Supported Response Formats

OpenAI Chat Completions

Total Context

1,000,000

Max Output

1,000,000

Input Price-40%

¥12 / 1M tokens

Output Price-40%

¥60 / 1M tokens

Cache Write-40%

¥15 / 1M tokens

Cache Read-40%

¥1.2 / 1M tokens

tokenmall-jdy

国内

TTFT

6.78s

Throughput

26.30tps

Uptime

100.00%

Provider Model

jdcloud/moonshotai/kimi-k3

Supported Parameters

temperaturetop_ptop_k

Recent Uptime

10月11日 11 PM100.00%

Reasoning

Toggleable

Supported Response Formats

OpenAI Chat Completions

Total Context

1,000,000

Max Output

1,000,000

Input Price

¥20 / 1M tokens

Output Price

¥100 / 1M tokens

Cache Write

¥25 / 1M tokens

Cache Read

¥2 / 1M tokens

tokenmall-zyld

国内

TTFT

20.13s

Throughput

16.58tps

Uptime

100.00%

Provider Model

tokenmall-zyld/moonshotai/kimi-k3

Supported Parameters

temperaturetop_ptop_k

Recent Uptime

10月11日 11 PM100.00%

Reasoning

Toggleable

Supported Response Formats

OpenAI Chat CompletionsOpenAI ResponsesAnthropic Messages

Total Context

1,000,000

Max Output

1,000,000

Input Price-43%

¥11.4 / 1M tokens

Output Price-43%

¥57 / 1M tokens

Cache Write-43%

¥14.25 / 1M tokens

Cache Read-43%

¥1.14 / 1M tokens

月

月之暗面

国内

TTFT

12.66s

Throughput

23.93tps

Uptime

100.00%

Provider Model

moonshot/moonshotai/kimi-k3

Supported Parameters

temperaturetop_ptop_k

Recent Uptime

10月11日 11 PM100.00%

Reasoning

Toggleable

Supported Response Formats

OpenAI Chat CompletionsAnthropic Messages

Total Context

1,000,000

Max Output

1,000,000

Input Price

¥20 / 1M tokens

Output Price

¥100 / 1M tokens

Cache Write

¥25 / 1M tokens

Cache Read

¥2 / 1M tokens

火

火山引擎

国内

TTFT

10.58s

Throughput

18.59tps

Uptime

100.00%

Provider Model

volcengine/moonshotai/kimi-k3

Supported Parameters

temperaturetop_ptop_k

Recent Uptime

10月11日 11 PM100.00%

Reasoning

Toggleable

Supported Response Formats

OpenAI Chat CompletionsOpenAI Responses

Total Context

1,000,000

Max Output

1,000,000

Input Price

¥20 / 1M tokens

Output Price

¥100 / 1M tokens

Cache Write

¥25 / 1M tokens

Cache Read

¥2 / 1M tokens

百

百度文心

国内

TTFT

26.25s

Throughput

14.44tps

Uptime

100.00%

Provider Model

baidu/moonshotai/kimi-k3

Supported Parameters

temperaturetop_ptop_k

Recent Uptime

10月11日 11 PM100.00%

Reasoning

Toggleable

Supported Response Formats

OpenAI Chat Completions

Total Context

1,000,000

Max Output

1,000,000

Input Price-10%

¥18 / 1M tokens

Output Price-10%

¥90 / 1M tokens

Cache Write-10%

¥22.5 / 1M tokens

Cache Read-10%

¥1.8 / 1M tokens

通

通义千问

国内

TTFT

7.11s

Throughput

31.90tps

Uptime

100.00%

Provider Model

alibaba/moonshotai/kimi-k3-2

Supported Parameters

temperaturetop_ptop_k

Recent Uptime

10月11日 10 PM100.00%

Reasoning

Toggleable

Supported Response Formats

OpenAI Chat CompletionsAnthropic Messages

Total Context

1,000,000

Max Output

1,000,000

Input Price

¥20 / 1M tokens

Output Price

¥100 / 1M tokens

Cache Write

¥25 / 1M tokens

Cache Read

¥2 / 1M tokens

通

通义千问

国内

TTFT

9.34s

Throughput

20.90tps

Uptime

100.00%

Provider Model

alibaba/moonshotai/kimi-k3

Supported Parameters

temperaturetop_ptop_k

Recent Uptime

10月11日 11 PM100.00%

Reasoning

Toggleable

Supported Response Formats

OpenAI Chat Completions

Total Context

1,000,000

Max Output

1,000,000

Input Price

¥20 / 1M tokens

Output Price

¥100 / 1M tokens

Cache Write

¥25 / 1M tokens

Cache Read

¥2 / 1M tokens

Performance for moonshotai/kimi-k3

Compare different providers across Zhinao API

Throughput

21.77 tok/s

TTFT

19.98 s

Uptime for moonshotai/kimi-k3

Uptime for moonshotai/kimi-k3 across all providers

Sample code and API for moonshotai/kimi-k3

Get API Key

Zhinao API normalizes requests and responses across providers for you

View full docs

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.360.cn/v1",
  apiKey: process.env.ZHINAO_API_KEY,
});

const response = await client.chat.completions.create({
  model: "moonshotai/kimi-k3",
  messages: [
    { role: "user", content: "Hello, how are you?" }
  ],
  temperature: 0.7,
  max_tokens: 1000,
});

console.log(response.choices[0].message.content);